Deep neural network-based fusion model for emotion recognition using visual data

The Journal of Supercomputing - In this study, we present a fusion model for emotion recognition based on visual data. The proposed model uses video information as its input and generates emotion...     

A new ECG beat clustering method based on kernelized fuzzy c-means and hybrid ant colony optimization for continuous domains

The kernelized fuzzy c-means algorithm uses kernel methods to improve the clustering performance of the well known fuzzy c-means algorithm by mapping a given dataset into a higher dimensional space non-linearly. Thus, the newly obtained dataset is more likely to be linearly seprable. However, to further improve the clustering performance, an optimization method is required to overcome the drawbacks of the traditional algorithms such as, sensitivity to initialization, trapping into local minima and lack of prior knowledge for optimum paramaters of the kernel functions. In this paper, to overcome these drawbacks, a new clustering method based on kernelized fuzzy c-means algorithm and a recently proposed ant based optimization algorithm, hybrid ant colony optimization for continuous domains, is proposed. The proposed method is applied to a dataset which is obtained from MIT–BIH arrhythmia database. The dataset consists of six types of ECG beats including, Normal Beat (N), Premature Ventricular Contraction (PVC), Fusion of Ventricular and Normal Beat (F), Artrial Premature Beat (A), Right Bundle Branch Block Beat (R) and Fusion of Paced and Normal Beat (f). Four time domain features are extracted for each beat type and training and test sets are formed. After several experiments it is observed that the proposed method outperforms the traditional fuzzy c-means and kernelized fuzzy c-means algorithms.     

Modal logic as metalogic

The goal of this paper is to show how modal logic may be conceived as recording the derived rules of a logical system in the system itself. This conception of modal logic was propounded by Dana Scott in the early seventies. Here, similar ideas are pursued in a context less classical than Scott's.First a family of propositional logical systems is considered, which is obtained by gradually adding structural rules to a variant of the nonassociative Lambek calculus. In this family one finds systems that correspond to the associative Lambek calculus, linear logic, relevant logics, BCK logic and intuitionistic logic. Above these basic systems, sequent systems parallel to the basic systems are constructed, which formalize various notions of derived rules for the basic systems. The deduction theorem is provable for the basic systems if, and only if, they are at least as strong as systems corresponding to linear logic, or BCK logic, depending on the language, and their deductive metalogic is not stronger than they are.However, though we do not always have the deduction theorem, we may always obtain a modal analogue of the deduction theorem for conservative modal extensions of the basic systems. Modal postulates which are necessary and sufficient for that are postulates of S4 plus modal postulates which mimic structural rules. For example, the modal postulates which Girard has recently considered in linear logic are necessary and sufficient for the modal analogue of the deduction theorem.All this may lead towards results about functional completeness in categories. When functional completeness, which is analogous to the deduction theorem, fails, we may perhaps envisage a modal analogue of functional completeness in a modal category, of which our original category is a full subcategory.     

APPLICATION OF THE CONSTANT MOLAR FLOW METHOD IN THE STUDY OF MASS TRANSFER OF GASES AND VAPORS IN BIDISPERSE STRUCTURED SOLIDS

In this paper, we provide exact analytical solutions for a new method of constant molar flow of pure gas into a reservoir containing solid adsorbents having a bidispersed structure, such as zeolite, activated carbon and alumina. Depending on the diffusion mechanism, two models are proposed under the isothermal conditions and linear isotherm. One is applicable to zeolite, and the other for alumina and activated carbon. The solutions of these two models are in the form of an infinite series, and its eigenvalues are determined from a transcendental equation. This transcendental equation is such that the eigenvalues are lumped into infinite but countable groups, and in each group there are infinite but countable eigenvalues. The dominant eigenvalues are generally the first ten to thirty eigenvalues (depending on the value of time at which the series is evaluated) in each group. The pressure response of the reservoir is a function of time as the result of the introduction of adsorbate into the reservoir, and as time is sufficiently large it asymptotes to a straight line, of which the slope is a function of the equilibrium parameters only and the intercept is found to be an explicit function of the equilibrium parameters as well as the diffusivities in the macropore and micropore.     

The structure of model coatings: latex-bound plastic pigment coatings

The immobilization and consolidation of model coatings based on monodisperse polystyrene (plastic pigment) and S/B latexes of known particle sizes were studied in terms of their packing volumes and the extent of latex shrinkage, which was found to increase with increasing pigment volume up to the critical pigment volume concentration (CPVC). The maximum latex shrinkage was estimated from the CPVC. Then, the porosity of model coatings was calculated based on three proposed latex shrinkage models: maximum, minimum, and linearly decreasing latex shrinkage. The number of pores and the average equivalent spherical pore diameters were subsequently calculated. The opacity and gloss of model coatings on polyester films were measured and their porosity was also determined by a simple coat weight-thickness method. The CPVC values determined by the opacity, gloss, and porosity versus PVC relationships, respectively, agreed very well with each other. The CPVCs determined by the opacity and porosity versus PVC curves were identical. The comparison between the theoretically calculated and experimental porosity values showed that the linearly decreasing value between the maximum and minimum latex shrinkage would best fit the experimental porosity data. The effect of plastic pigment particle size on the optical properties and porosity of model coatings was also studied and it was observed that the coating opacity and porosity increased with increasing plastic pigment particle size, but the gloss decreased. Additionally, a minimum crack-free temperature (MCFT) of latex-bound coatings was proposed to better predict the behaviors of latexes as pigment binders. The wet state of model coating dispersions, the surfaces of consolidated model coatings, and their internal structure were examined by both electron and atomic force microscopy, and their micrographs were found to be consistent with our immobilization and consolidation models.     

Development of bond strength model for FRP plates using back‐propagation algorithm

For the flexural reinforcement of bridge and building structure, synthetic materials whose dynamic properties are superior and those containing the merit of corrosion‐proof are widely used as the substitute for a steel plate. Since FRP plate has improved bond strength owing to the fibers externally adhering to the plate, many researches regarding the bond strength improvement have been substantially performed. To search out such bond strength improvement, previous researchers had ever examined the bond strength of FRP plate through their experiment by setting up many variables. However, since the experiment for a research on the bond strength takes much of expenditure for setting up the equipment and is time‐consuming, also is difficult to be carried out, it is limitedly conducted. The purpose of this study was to develop the most suitable artificial neural network model by application of various neural network models and algorithm to the data of the bond strength experiment conducted by previous researchers. Many variables were used as input layers against bond strength: depth, width, modulus of elasticity, tensile strength of FRP plate and the compressive strength, tensile strength, and width of concrete. The developed artificial neural network model has been applied back‐propagation, and its error was learned to be converged within the range of 0.001. Besides, the process for the over‐fitting problem has been dissolved by Bayesian technique. The verification on the developed model was executed by comparison with the test results of bond strength made by other previous researchers, which was never been utilized to the learning as yet. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 100: 5119–5127, 2006     

New Process for the Preparation of Monodispersed, Spherical Silica Particles

A new method is presented for preparing highly monodispersed silica particles using a two-stage semibatch/batch hydrolysis reaction of Si(OC₂H₅)₄. The slower rate of hydrolysis of the tetraethylorthosilicate (TEOS) that occurred during the semibatch process resulted in larger silica particles with a higher yield and narrower size distribution. This was in direct contrast to the batch process. In addition, the ability of four different mixed processes to produce silica particles with good packing density, narrower particle-size distribution, and higher yield were evaluated. These were batch/batch (B-B), batch/semibatch (B-S), semibatch/batch (S-B), and semibatch/semibatch (S-S) processes. The S-S system produced the largest particles with the highest yields. The size of the silica particles obtained by the S-B method decreased with increasing reaction time, while the particles obtained by the B-S process had the best particle-size distribution and packing density. In conclusion, a mixed batch/semibatch system was the best way to produce an extremely narrow particle-size distribution and a good packing density.     

Flaw-Tolerance and R-Curve Behavior of Liquid-Phase-Sintered Silicon Carbides with Different Microstructures

β-SiC powder containing 6 wt% A1₂O₃ and 4 wt% Y₂O₃ as sintering additives was pressureless sintered at 2000°C for 1 h (AYE-SiC) and 3 h (AYP-SiC). AYE-SiC consisted of an equiaxed grain structure and AYP-SiC exhibited a micro-structure with platelike grains as a result of grain growth related to β→α phase transformation during sintering, R -curve behavior and flaw tolerance for these silicon carbides were evaluated by the indentation-strength technique. For comparison, the R -curve behavior of conventional sintered, boron- and carbon-doped SiC (SS-SiC) was evaluated. AYE-SiC and AYP-SiC exhibited rising R -curve behavior with toughening exponents of m = 0.042 and m = 0.135, respectively. AYP-SiC exhibited better flaw tolerance and more sharply rising R -curve behavior than AYE-SiC. The more sharply rising R -curve behavior and the better flaw tolerance of AYP-SiC were attributed mainly to grain bridging of crack faces by platelike grains. Because of the high degree of transgranular fracture, SS-SiC exhibited a flat R -curve despite a microstructural feature with platelike grains.     

Low‐temperature photoinitiation solution polymerization behavior of N‐vinylcarbazole in tetrahydrofuran

N‐Vinylcarbazole (VCZ) was solution‐polymerized in tetrahydrofuran (THF) at ?20, 0, and 20°C using the photoinitiation method; the effects of the amount of solvent, polymerization temperature, and photoinitiator concentration were investigated. On the whole, the experimental results corresponded to predicted ones. Low polymerization temperature using photoinitiation proved to be successful in obtaining poly(N‐vinylcarbazole) (PVCZ) of a high molecular weight with a smaller temperature rise during polymerization; nevertheless of free radical polymerization by 2,2′‐azobis(2,4‐dimethylvaleronitrile) (ADMVN). The photo‐solution polymerization rate of VCZ in THF was proportional to the 0.47 power of ADMVN concentration. The molecular weight was higher and the molecular weight distribution was narrower with PVCZ polymerized at lower temperatures. For PVCZ prepared in THF at ?20°C using a photoinitiator concentration of 0.00005 mol/mol of VCZ, a weight‐average molecular weight of 510,000 was obtained, with a polydispersity index of 1.73, and a degree of lightness converged to about 99%. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 86: 3667–3672, 2002     

Manufacturing and foaming of high melt viscosity of polypropylene by using electron beam radiation technology

The high melt viscosity of polypropylene was studied by grafting bifunctional monomers, 1,6‐hexanediol diacrylate (HDDA) and tripropylene glycol diacrylate (TPGDA), onto homopolypropylene (HPP) and random ter‐polypropylene (RTPP) under electron‐beam irradiation. Creation of the high‐melt‐viscosity polypropylene was possible at low radiation dosage and low monomer content, under a prohibition of both radiation degradation and homopolymerization. TPGDA monomer was more effective in increasing the melt viscosity of HPP compared with RTPP, whereas HDDA monomer was more effective for enhancing the melt viscosity of RTPP. Such different effects of monomers on melt viscosity may arise from different monomer structures, namely, TPGDA has additional three methyl groups, but HDDA has no methyl groups. Electron‐beam radiation technology, on an increase of the melt viscosity, was much more effective in HPP than RTPP, when compared with virgin polymers. Modified RTPP and HPP with high melt viscosity were capable of foaming with numerous fine cells, of which the modified HPP with 1.5 mmol TPGDA and 0.5 kGy could create more spherical foam cells and its bending strength was 1.5 times more than that of the foamed RTPP. POLYM. ENG. SCI., 46:431–437, 2006. © 2006 Society of Plastics Engineers.